Lately I’ve been
realizing how much machine learning (ML) and artificial intelligence can be applied to our sports culture. If you read my 2017 predictions blog, you’ll know that I see major implications for ML in the coming years. So let’s take a look at how we use technology in sports and fan engagement, and how that can translate to the enterprise.
Scouting in sports is a human pursuit. It takes a highly skilled, observant and excellent judge to watch an athlete perform at an amateur level, and understand if he or she has what it takes to go pro. But what if a machine could do that just as well – if not better – than a human? With wearable technologies, data is available to show us every bit of relevant information about an athlete – heart rate, muscle mass, bone density…the list goes on. By analyzing that information, compiled with broader data, like current team lineup and weaknesses, a machine could conceivably review dozens of athletes to determine which would make the best fit in the organization. You could argue that this is still a field that relies too much on gut instinct to ever fully transform to a machine. But it’s not impossible to imagine machine learning playing a much bigger role in recruitment and team dynamics in the future.
How does this apply to the enterprise? Take hiring. Even the fairest, most judicious hiring manager may be prone to biases he or she is unaware of. But software now exists to remove that bias entirely. HCM technology can address gender equity in staffing, management, development, retention, compensation and promotion, overcoming inequalities that humans create. Will the machines soon make all hiring decisions? Again, a gut instinct cannot be replicated – but it can be used in tandem.
Any major sports fan knows the feeling: your team has scored, but a flag was thrown. Did the ball cross the goal line? Did the receiver make a catch with both feet in bounds? Fans watch with anticipation as the officials confer, and confer, and confer before they make a decision. One official may see the two feet touching ground, and the other may vehemently disagree. Even having high definition cameras and millisecond by millisecond playback doesn’t change what the human eye sees. But with the aforementioned wearable technology, as well as sensors that can be placed on the turf itself, the opinion goes by the wayside. There’s no debate about where the senor detected movement because the sensor cannot lie, and a sensor isn’t biased.
Some officials in baseball already resist this. As shown in an episode of HBO’s Real Sports, after determining they only call throws correctly 88 percent of the time, umpires were given a disc after every game to review which throws were called incorrectly. One longtime umpire’s response was to throw the disc away without even watching it!
How does that relate to the enterprise? The guesswork can be completely eliminated in operations. In a manufacturing plant, there may be a lengthy assembly line to create a product. Suddenly, the product has a flaw. One overseer think it’s Machine A that’s malfunctioning, and his colleague thinks it’s Machine B (or Part A and Part B…you get the picture). Or maybe every machine or part needs to be inspected. Not anymore. With sensory data, the exact issue can be identified nearly instantly. And with machine learning, the machine can identify what issue led to the problem, and suggest ways to fix it so it’s avoided in the future.
Second in hype only to the game itself are the Super Bowl commercials. We laugh at some, we jeer at others, and the special ones are remembered for years to come. But not all commercials serve the same audience. Some things are universally appreciated, like snack food. But what about the ads not targeted to you? For example, a woman isn’t the demographic audience for men’s deodorant. With DVRs, we quickly fast forward through those ads we don’t care about. A smart machine will, over time, understand what the user wants to see, and show only that content. Advertisers may hate the idea at first but they can use this to their advantage. Maybe a woman will fast forward a men’s deodorant ad…unless that ad stars Ryan Gosling.
This example applies to data in the enterprise as a whole. Every organization is managing seas of data and they will soon become oceans. You have to make sense of it, separate the good from the bad or unimportant, and analyze it to be competitive. Don’t waste time selling to someone who isn’t interested. Likewise, identify trends and ways you can make them interested!
How do you see machine learning applied to sports? Do you think it’s for the better, or should we leave things as they are?